CN107608936A - A kind of epicyclic gearbox combined failure feature extracting method - Google Patents
A kind of epicyclic gearbox combined failure feature extracting method Download PDFInfo
- Publication number
- CN107608936A CN107608936A CN201710868167.3A CN201710868167A CN107608936A CN 107608936 A CN107608936 A CN 107608936A CN 201710868167 A CN201710868167 A CN 201710868167A CN 107608936 A CN107608936 A CN 107608936A
- Authority
- CN
- China
- Prior art keywords
- mrow
- epicyclic gearbox
- ultiwavelet
- function
- wavelet
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The present invention discloses a kind of epicyclic gearbox combined failure feature extracting method, first, measures and stores epicyclic gearbox vibration signal；Secondly, the symmetrical lift frame of m ultiwavelet is constructed, introduces regulation and control parameter；Then, evaluation index of the multi-fractal entropy as Adaptive matching criterion is built, the self-adaptive construction of m ultiwavelet is carried out by intelligent optimization algorithm, obtains the multi-wavelet bases function to match with Dynamic Signal；Decomposed again by redundancy multi-wavelet transformation；Finally, the relative energy ratio at the fault characteristic frequency in each frequency range is calculated, frequency band relative energy is obtained than block diagram, selects Fault-Sensitive frequency band, and then identify and isolate combined failure.Epicyclic gearbox bang path can be overcome complicated by the present invention and operating mode influence of noise, and by adaptive m ultiwavelet construction and sensitive features frequency band selection, epicyclic gearbox ring gear, planetary gear and sun gear fault features are isolated in extraction.
Description
Technical field
The present invention relates to epicyclic gearbox fault detection technique field, and in particular to a kind of epicyclic gearbox combined failure is special
Levy extracting method.
Background technology
Epicyclic gearbox small volume, in light weight, gearratio are big, efficiency high, bearing capacity are strong, it may have more power convergences or single
The advantages of power is disperseed, it is widely used in the industry-by-industry system such as new-energy automobile, wind-driven generator, Aero-Space, ship.By
Often run in epicyclic gearbox under the severe environmental conditions such as high speed, heavy duty and thump, sun gear, planet in epicyclic gearbox
With ring gear etc. crucial zero the various faults such as abrasion, fatigue, broken teeth and crackle easily occur for wheel, and further induce other events
Barrier, so as to cause huge economic losses.Therefore, the running status of epicyclic gearbox is monitored and identifies its generation in time
Failure, there is important engineering significance.
Epicyclic gearbox includes sun gear, planetary gear and ring gear, in running sun gear simultaneously with multiple planetary gears
Engagement, it is easiest to that fatigue and Crack Damage occurs.And multipair gear mesh engages in epicyclic gearbox simultaneously, planetary gear position is continuous
Change causes the continuous change of the engagement force position and direction of gear mesh, and signal drive path constantly changes, and causes planetary gear
Case is that one simple in construction and failure mechanism and the complicated mechanical system of spectrum structure, its troubleshooting issue and traditional dead axle tooth
Roller box, which is compared, has the various features and difficult point of its own.Conventional gearbox Troubleshooting Theory can not be solved effectively with technology at present
Many problems that certainly Identification of Cracks of epicyclic gearbox is faced.
The characteristics of for planetary gear box fault diagnosis and difficult point, need to be by effective fault diagnosis technology and side
Method, accurate extraction fault signature, the purpose of accurate quantitative analysis tracing trouble can be reached.Wavelet transformation is described as " school microscop "
It is to handle the powerful that Weak characteristic extracts in non-stationary signal.Its physical essence is to seek in signal to include and " basic function "
Most like or maximally related component.But it only has a basic function, inborn deficiency in terms of Trouble Match be present, and it is how small
New development of the ripple as small echo, it has both a variety of advantageous properties that single wavelet can not be provided simultaneously with, and when possessing multiple simultaneously
The basic function that frequency feature has differences so that m ultiwavelet has notable in terms of crackle Weak characteristic and MULTIPLE COMPOSITE feature extraction
Advantage.But fixed multi-wavelet bases function can not still be realized and planetary transmission system crackle is faint or MULTIPLE COMPOSITE feature
Optimum Matching, limit its ability orientation in the extraction of planetary transmission system crack.
The content of the invention
To be solved by this invention is that conventional gearbox method for diagnosing faults can not effectively solve the crackle of epicyclic gearbox
Identify problem encountered, there is provided a kind of epicyclic gearbox combined failure feature extracting method.
To solve the above problems, the present invention is achieved by the following technical solutions：
A kind of epicyclic gearbox combined failure feature extracting method, including step are as follows：
Step 1, epicyclic gearbox vibration signal picked up using acceleration vibrating sensor, the acceleration transducer is arranged on
Epicyclic gearbox input shaft end to be measured covers；
Step 2, the self-adaptive construction that multi-wavelet bases function is carried out to the vibration signal collected；
Step 2.1, select initial m ultiwavelet and other wavelet basis functions and multi-scaling Functions for correcting m ultiwavelet
Translational movement, using symmetric condition and vanish-moment construction lifting system of linear equations, solve owe fixed condition under linear equation
Group obtains Lifting Coefficients, introduces regulatable free parameter in the process, and Lifting Coefficients substitution Lifting Coefficients equation is gone forward side by side
Row transform realizes the symmetrical lifting of m ultiwavelet, completes the construction of multi-wavelet bases function library；
How small step 2.2, the design feature according to epicyclic gearbox fault signature, construction normalization multi-fractal entropy conduct be
Evaluation index during ripple basic function adaptive optimization, obtained and adaptable optimal of fault signature based on intelligent optimization algorithm
Multi-wavelet bases function；
Step 3, the signal gathered using optimal multi-wavelet bases function pair carry out redundancy multi-wavelet transformation, obtain multiple points
Sub-band after solution；
Step 4, relative energy ratio at the fault characteristic frequency of each sub-band is calculated, by the higher son of relative energy
Frequency band is as the sensitive sub-band where combined failure feature；
Step 5, Hilbert envelope demodulation processing is carried out to the sensitive sub-band that decomposition obtains one by one, extract planet tooth
Roller box combined failure correlated characteristic, and carry out identifying and diagnosing.
In above-mentioned steps 2.1, free parameter is determined by solving the Lifting Coefficients non trivial solution owed under fixed condition.
The detailed process of above-mentioned steps 2.2 is as follows：
First, morther wavelet of the selection with suitable vanishing moment, determines partition function, and describe partition letter by quality factor
Several Scaling behaviors, and then Legendre is carried out to quality factor and converts to obtain the multifractal spectra of signal；
Secondly, comentropy is calculated and be introduced into multifractal spectra, construct multi-fractal entropy, and be normalized to obtain
Normalize multi-fractal entropy；
Finally, based on intelligent optimization algorithm, returning for detail signal corresponding to latter two wavelet basis function is decomposed with m ultiwavelet
The one change minimum optimization object function of multi-fractal entropy sum optimizes, and obtains adaptive multi-wavelet bases function.
In above-mentioned steps 2.2, intelligent optimization algorithm is genetic algorithm.
In above-mentioned steps 2.2, multi-fractal entropy H is normalizednFor：
In formula, f (αi) be signal multifractal spectra.
In above-mentioned steps 3, the number of plies that redundancy multi-wavelet transformation is carried out to multi-wavelet bases function is 3 layers.
In above-mentioned steps 4, the relative energy at the fault characteristic frequency of each sub-band is than r：
Wherein fc∈(fc- Δ, fc+Δ)
In formula, A be squared envelope spectrum amplitude, fcFrequency is characterized, Δ is selected frequency separation, and f=0~f ' is
Sub-band scope.
Two or more different faults features, and the failure of different faults type are included for epicyclic gearbox combined failure
The characteristics of signature waveform difference, the present invention have multiple wavelet basis functions using m ultiwavelet, had in combined failure feature extraction
There is inherent advantage, while avoid fixed multi-wavelet bases function from can not realizing best match with fault signature waveform, be unfavorable for event
Hinder the optimum extraction of feature, propose the adaptive m ultiwavelet building method based on multi-fractal entropy, this method possesses following notable
Advantage：
1. instant invention overcomes epicyclic gearbox combined failure to diagnose problem, there are multiple wavelet basis functions using m ultiwavelet
The characteristics of, multiple wavelet basis functions are matched with multiple fault signatures so that the feature extraction of epicyclic gearbox combined failure with
It is separated into possibility；
What 2. the present invention caught plant equipment local anomaly using fractals and induced has geometry feature not
Regular Singularity Signal, normalization multi-fractal entropy is constructed as optimizing index, the normalization after being decomposed using m ultiwavelet is more
Divide the optimal multi-wavelet bases function that shape entropy sum minimum principle is guided and optimization is adapted with combined failure feature again, obtain excellent
Good combined failure feature extraction and separating capacity；
3. at the fault characteristic frequency in each frequency range after being decomposed by m ultiwavelet relative energy than calculating, obtain frequency
Band relative energy intuitively shows the fault signature energy of each parts of epicyclic gearbox, is advantageous to sensitivity than block diagram
The selection of characteristic frequency；
4. the present invention can be used for the planetary gear box fault diagnosis based on vibration monitoring, sun gear, ring gear can be extracted
With the early stage combined failure feature of planetary gear, avoid the sudden accident of transmission system from occurring, reduce economic loss.
Brief description of the drawings
Fig. 1 is a kind of diagnostic method flow chart of epicyclic gearbox combined failure；
Fig. 2 is epicyclic gearbox vibration signal time domain beamformer；
Fig. 3 is epicyclic gearbox original signal spectrum figure；
Fig. 4 is epicyclic gearbox original signal envelope spectrogram；
Fig. 5 is the adaptive multi-wavelet bases function of epicyclic gearbox vibration signal；Wherein (a) is m ultiwavelet function ψ1, (b) is
M ultiwavelet function ψ2；
Fig. 6 is that epicyclic gearbox multi-wavelet packets decompose energy ratio block diagram；
Fig. 7 is the envelope spectrogram of sensitive bands branch；Wherein (a) is the 7th branch's envelope spectrum, and (b) is the 16th branch's envelope
Spectrum.
Embodiment
For the object, technical solutions and advantages of the present invention are more clearly understood, below in conjunction with instantiation, and with reference to attached
Figure, the present invention is described in more detail.
A kind of epicyclic gearbox combined failure feature extracting method, as shown in figure 1, it specifically comprises the following steps：
The first step：Signal acquisition.
Epicyclic gearbox vibration signal is picked up using vibration acceleration sensor, the acceleration is arranged on planetary gear to be measured
Case input shaft end covers.
Second step：The self-adaptive construction of multi-wavelet bases function is carried out to the vibration signal collected.
First, multi-wavelet bases construction of function is carried out using the symmetrical lift frame of m ultiwavelet, ensures to filter by symmetry
The linear phase of utensil or generalizing linear phase, when being advantageous to BORDER PROCESSING, while avoiding decomposing signal and being reconstructed
Phase distortion；Regulatable free parameter is introduced in the process；Secondly, it is special according to the structure of epicyclic gearbox fault signature
Point, construction normalization multi-fractal entropy are excellent based on intelligence as the evaluation index during multi-wavelet bases function adaptive optimization
Change algorithm and obtain the multi-wavelet bases function being adapted with combined failure feature.
(1) multi-wavelet bases construction of function is carried out using lift frame.
First, initial m ultiwavelet ω is selected0(x), wherein ω0(x)=ψ1Or ψ2；Then select for correcting m ultiwavelet
Other basic function ω1(x),...,ωk(x) translational movement k.Finally new m ultiwavelet can be constructed by " Lifting Coefficients equation "Lifting Coefficients equation is：
M ultiwavelet has more advantages compared with single wavelet on lifting construction, in being lifted such as single wavelet, for correcting
Original wavelet function can only be scaling function, i.e. ω in above formulaiIt is onlyAnd in m ultiwavelet lifting, it is a certain for correcting
M ultiwavelet function not only includes two multi-scaling Functions, can also be another corresponding m ultiwavelet function, i.e., for ψ1,ForObviously, will for constructing the basic function of new m ultiwavelet function
More than single wavelet, come the bigger free degree and flexibility for the structural belt of m ultiwavelet, to meet more, more specifically to require.
Assuming that the vanishing moment of m ultiwavelet is promoted to p ' by p, " Lifting Coefficients equation " both sides are integrated, can obtained down
The lifting system of linear equations in face.
Solution of equations { ciBe m ultiwavelet lifting function coefficient.M ultiwavelet can be obtained to formula (1) progress transform to carry
Rise framework.
(2) the symmetrical lifting of m ultiwavelet is carried out.
To ensure the symmetry of lifting process, " symmetrical selection " method is utilized to select other letters for correcting m ultiwavelet
Several translational movements.Assuming that initial multi-scaling FunctionsWith m ultiwavelet function ψ1、ψ2To be symmetrical or antisymmetric, in symmetrical
The heart is respectivelyThen symmetrical method for improving is expressed as below, with ψ1Symmetrical lifting exemplified by, lifting function it is flat
Shifting amountIt must meet：
In formula：I=1,2；J=1,2 ...；M=1,2.
OrderThe symmetry of initial multi-scaling Functions and initial m ultiwavelet function is represented respectively
Matter, wherein 1 represent symmetry, -1 make difficulties title property.WillWith M (ψi, k, n) and=∫ ψi
(x+k)xnDx substitutes into formula (2), and equal sign first left matrix is represented into MB, wherein MB=MB, M are respectively with B
It is symmetry matrix to make B
Coefficient vector in formula (2) is expressed asAnd it is expressed as M on the right of equationψ=[M
(ψi,0,p),M(ψi,0,p+1),…M(ψi,0,p'-1)]T, then formula (2) be changed into following formula
MBC=Mψ (6)
Non trivial solution C is to be used to lift ψ1Coefficient, ψ2Situation it is similar therewith, only difference is that lifting ψ2's
Function isWithLifting Coefficients are substituted into formula (2), and carries out transform and obtains lifting matrixes T and S.Therefore, m ultiwavelet is symmetrical
Lifting construction can be realized by means of lifting matrixes T and S, specific as follows：
In formula：HnewFor the low pass filter symbol after the symmetrical lifting construction of m ultiwavelet；GnewStructure is symmetrically lifted for m ultiwavelet
High-pass filter symbol after making.
(3) free parameter obtains in self-adaptive construction.
Three kinds of situations be present in the solution of Lifting Coefficients equation (6)：(a) equation is overdetermination, no solution；(b) equation is positive definite, only
One solution；(c) equation is owes fixed, more solutions.Flexibility using method for improving construction m ultiwavelet comes from lifting process with the free degree
In free parameter, this requires that non trivial solution must be the third.To the matrix M in formula (6)BAbout subtracted, former linear equation
Group can be reduced to irredundant system of linear equations, and equation number is Rank (MB), Rank refers to rank of matrix, therefore this is linear
Equation group may have Nf=(p '-p)-Rank (MB) individual free parameter.Because adaptive essence is by certain optimization method pair
The optimization process that contained free parameter is carried out according to specified target, it is seen that these free parameters are that m ultiwavelet realizes adaptive pass
Key.
(4) optimization aim constructs.
The signal that plant equipment is induced due to local anomaly often has singularity, and it shows as mutation, cusp etc. no
The transient assemblies of rule.And multi-fractal is defined in Fractal Non-Scale section, had the singular measure of multiple scaling exponents
The set formed, what it was portrayed is distribution situation of the fractal measure in support, and it is different can to describe point shape with spectral function
The feature of level.And in order to identify fault type and quantification degree of injury exactly, by comentropy and more points of shape spectral functions
It is combined, forms multi-fractal entropy.Adaptive m ultiwavelet construction is introduced using multi-fractal entropy as the object function of optimization process,
Multi-fractal entropy is calculated as follows：
If f (x) is the function of a finite energy, i.e. f (x) ∈ L2(R), then the wavelet transformation of the function is defined as follows formula institute
Show：
In formula：ψa,b(x) it is morther wavelet.
As yardstick a=a0When, if to b=b0A certain field in any point b have：
|Wf(a0,b；ψ)|≤|Wf(a0,b0；ψ)| (9)
Then claim (a0,b0) it is local modulus maxim point, and the line of all Maximum entropy principle points in metric space is referred to as Maximum entropy principle
Line.Contain abundant signal message in the size of wavelet transformation local modulus maxim and position.One important small echo local pole
The property of big loft is when analyzed signal f (x) is in certain point x0Hausdorff indexes α be less than small echo vanishing moment when, then
The point at least is pointed in the presence of a local maximum loft, and following Scaling behavior be present along very big loft wavelet conversion coefficient：
The fractal characteristic of signal is mainly shown as the level distribution of singularity, and can be protruded by small echo local maximum loft
The hierarchical structure of singularity.
With Wavelet modulus maxima come to calculate the key of multifractal spectra be representing for partition function, if below yardstick a's is all
The collection of Wavelet modulus maxima line is combined into L (a), then closing definition partition function in the collection is：
The supremum of Wavelet Modulus Maxima is selected by above formula, is overcome due to putting neighbouring wavelet transformation on maximum line
Caused Z (q, a) instability problem during modulus value very little.It also avoid simultaneously due to Z (q, a) maximum caused by quick concussion draw
The data sharp increase problem of hair.But during scaling function a → 0, the Scaling behavior of partition function can be by quality factor τ (q) come table
Reach
Z (q, a)=aτ(q) (12)
Convert to obtain by carrying out Legendre to quality factor τ (q)：
It is so as to obtain multifractal spectra f (α)：
If f (αi), 1≤i≤k be signal multifractal spectra, f (αi) size reflect point under corresponding singular index
Shape dimension proportion shared in total fractal dimension.According to the measure formulas of entropy, then there is multi-fractal entropy：
In formula：
Normalize multi-fractal entropy HnDefinition be：
From above formula, Hn∈[0,1].When device fails, different degrees of singularity, failure occurs in signal
Impact is more sharp, and its singular value is smaller；And the measurement of entropy degree of being to determine and complexity, the signal period property that singularity be present are got over
By force, when self-similarity is better, show that the state of equipment more determines, now, normalization multi-fractal entropy HnIt is smaller, closer to 0.
(5) self-adaptive construction.
In view of genetic algorithm has very strong robustness and the overall situation, parallel search feature, and object function is not needed
Mathematical expression between variable.The present invention uses genetic algorithm as optimized algorithm, the normalization multi-fractal entropy conduct of construction
Optimizing evaluation index, the normalization multi-fractal that m ultiwavelet decomposes detail signal corresponding to latter two wavelet basis function is obtained respectively
Entropy；The optimal selection problem of adaptive multi-wavelet bases function is converted into the letter of the details after two multi-wavelet bases function decompositions of object function
The minimum optimization problem of multi-fractal entropy sum is normalized corresponding to number；Acquisition is adapted optimal how small with combined failure feature
Ripple basic function.
3rd step：Sensitive features frequency band obtains.
By redundancy multi-wavelet transformation, the sub-band after multiple decomposition is obtained.Calculate the fault signature frequency of each sub-band
Relative energy ratio at rate, pass through relative energy obtain combined failure where frequency band more directly perceived than block diagram.
Using optimal multi-wavelet bases function 22 are produced after 3 layers of redundancy multi-wavelet packets decompose3=16 branches；Simultaneously
Give up m ultiwavelet restructuring procedure so as to obtain the information representation of multiple-limb, while by every after formula (17) calculating m ultiwavelet decomposition
Relative energy ratio at fault characteristic frequency in one frequency range, frequency band relative energy is obtained than block diagram, each multi-wavelet bases letter
Number decomposites 8 sub-bands, totally 16 sub-bands；Optimal sensitive bands are searched out in the distribution situation of faults from figure to enter
Row analysis.
Wherein fc∈(fc-Δ,fc+Δ) (17)
In formula：fcIt is characterized frequency/Hz；A is the amplitude of squared envelope spectrum；Δ is selected frequency separation, f=0~
F ' is frequency band range.
4th step：Combined failure feature extraction：
Hilbert envelope demodulation process is carried out to the sensitive sub-band that decomposition obtains one by one, epicyclic gearbox is extracted and answers
Close failure correlated characteristic and carry out identifying and diagnosing.
Below by an instantiation, the performance of the present invention is further elaborated：
(1) it is arranged on and treats using acceleration vibrating sensor pickup epicyclic gearbox vibration signal, the acceleration transducer
Epicyclic gearbox input shaft end is surveyed to cover.
The parameter for testing epicyclic gearbox is as shown in table 1.
The epicyclic gearbox parameter of table 1
The fault characteristic frequency calculation formula that epicyclic gearbox is respectively driven component is as follows：
In formula：finTurn frequency/Hz for input shaft；fsFor sun gear；fpFor fault characteristic frequency/Hz of planetary gear；frTo be interior
Fault characteristic frequency/Hz of gear ring；ZsFor the number of teeth of sun gear；ZpFor the number of teeth of planetary gear；ZrFor the number of teeth of ring gear.
During test, epicyclic gearbox input speed is 1200r/min, is accelerated in the input arrangement vibration of epicyclic gearbox
Sensor is spent, obtains vibration signal, sample frequency 12.8KHz.Calculated through formula (18)-(20), the epicyclic gearbox first order
Characteristic frequency is as shown in table 2.
The epicyclic gearbox first order drive characteristics frequency when rotating speed of table 2 is 1200r/min
The vibration signal time domain waveform that vibration acceleration sensor is gathered is as shown in Fig. 2 data length is 32768.Make
The frequency spectrum and Hilbert obtained with traditional FFT methods converts obtained envelope spectrum as shown in Figure 3 and Figure 4.
As can be seen from Figure 2 time-domain signal is flooded by noise substantially, it is difficult to judges fault signature.From its spectrogram 3
The upper energy that can see is concentrated mainly in meshing frequency and its frequency multiplication of primary planet pinion case, and its meshing frequency is nearby entered
Row amplification, it can be seen that the sideband using ring gear fault characteristic frequency as interval be present.From envelope spectrum low-frequency range enlarged drawing 4
It can be seen that the fault characteristic frequency of ring gear is still obvious.Therefore, infer that the epicyclic gearbox there may be ring gear failure,
But the other fault signatures of epicyclic gearbox in original signal spectrum and envelope spectrum are flooded by noise.
(2) it is initial m ultiwavelet to select Hermite, and m ultiwavelet vanishing moment is lifted to 5 ranks, entered using symmetrical lift frame
Row m ultiwavelet is lifted, and obtains free parameter, minimum excellent with the normalization multiple analysis entropy sum after decomposing by genetic algorithm
Change target, obtain optimal multi-wavelet bases function as shown in figure 5, Fig. 5 (a) is m ultiwavelet function ψ1, Fig. 5 (b) is m ultiwavelet function
ψ2。
(3) because the diversity and complexity of epicyclic gearbox combined failure feature, its fault signature may be distributed in not
Same time-frequency location, in order to comprehensively, intactly embody the analysis result of combined failure, first, use optimal multi-wavelet bases function
22 are produced after three layers of redundancy multi-wavelet packets decompose3=16 branches；Secondly, m ultiwavelet restructuring procedure is given up so as to obtain
The information representation of multiple-limb, the relative energy ratio at the fault characteristic frequency in each frequency range after m ultiwavelet decomposes is calculated, is obtained
Frequency band relative energy is obtained than block diagram, as shown in fig. 6, preceding 8 branches correspond to the frequency of first multi-wavelet bases function decomposition in figure
Band, the frequency band of second multi-wavelet bases function decomposition of rear 8 correspondences, searched out from figure in the distribution situation of faults optimal
Sensitive bands are analyzed.
It is seen from figure 6 that planetary gear failure and sun gear failure relative energy concentrate on the 7th branch, and ring gear failure
Relative energy up to the 16th branch.The 7th branch and the 16th branch is selected to do Hilbert envelope spectrums to it as sensitive bands,
As shown in Figure 7.It can see from Fig. 7 (a), planetary gear wheel fault characteristic frequency 5.83Hz and sun gear fault characteristic frequency
52.5Hz is protruded, while also includes ring gear fault characteristic frequency 7.5Hz.Then clearly see in Fig. 7 (b), ring gear event
The composition for hindering characteristic frequency 7.5Hz is very prominent.Ring gear can determine whether there occurs damage fault by Fig. 7, and planetary gear and too
Sun wheel then generates smaller damage fault.
OOBA finds there is a 7 × 0.5mm on ring gear2Cut, on planetary gear there is also one 3 ×
0.5mm2Cut and a 2 × 1mm2Attrition fault, and on sun gear there are some pitting faults in this.So as to demonstrate this
Validity of the method in epicyclic gearbox combined failure feature extraction.
The present invention proposes a kind of epicyclic gearbox combined failure feature extracting method.First, measure and store planetary gear
Case vibration signal；Secondly, the symmetrical lift frame of m ultiwavelet is constructed, introduces regulation and control parameter；Then, structure normalization multi-fractal entropy
As the evaluation index of Adaptive matching criterion, the self-adaptive construction of m ultiwavelet is carried out by intelligent optimization algorithm, is obtained and dynamic
The multi-wavelet bases function that state signal matches；Decomposed again by redundancy multi-wavelet transformation；Finally, the failure in each frequency range is calculated
Relative energy ratio at characteristic frequency, frequency band relative energy is obtained than block diagram, selects Fault-Sensitive frequency band, and then identify and divide
Separate out combined failure.
The present invention is efficiently and reliably decomposed by simple vibration measurement using self-adaptive redundant multi-wavelet transformation
Multiple sub-band signals are obtained, and then the relative energy at the fault characteristic frequency for passing through each sub-band is than obtaining fault signature
Sensitive bands, pass through Hilbert conversion extraction epicyclic gearbox correlated characteristics and identifying and diagnosing.The present invention can overcome planet
Gear-box bang path complexity and operating mode influence of noise, by adaptive m ultiwavelet construction and sensitive features frequency band selection, extraction
Isolate epicyclic gearbox ring gear, planetary gear and sun gear fault features.
It should be noted that although embodiment of the present invention is illustrative above, but it is to the present invention that this, which is not,
Limitation, therefore the invention is not limited in above-mentioned embodiment.Without departing from the principles of the present invention, it is every
The other embodiment that those skilled in the art obtain under the enlightenment of the present invention, it is accordingly to be regarded as within the protection of the present invention.
Claims (7)
1. a kind of epicyclic gearbox combined failure feature extracting method, its feature are being that it specifically comprises the following steps：
Step 1, epicyclic gearbox vibration signal is picked up using acceleration vibrating sensor, the acceleration transducer is arranged on to be measured
Epicyclic gearbox input shaft end covers；
Step 2, the self-adaptive construction that multi-wavelet bases function is carried out to the vibration signal collected；
Step 2.1, select initial m ultiwavelet and for correct m ultiwavelet other wavelet basis functions and multi-scaling Functions it is flat
Shifting amount, using symmetric condition and vanish-moment construction lifting system of linear equations, solve the system of linear equations owed under fixed condition and obtain
Lifting Coefficients are obtained, introduce regulatable free parameter in the process, Lifting Coefficients are substituted into Lifting Coefficients equation and carry out Z changes
The symmetrical lifting for realizing m ultiwavelet is changed, completes the construction of multi-wavelet bases function library；
Step 2.2, the design feature according to epicyclic gearbox fault signature, construction normalization multi-fractal entropy is as multi-wavelet bases
Evaluation index during function adaptive optimization, obtained and adaptable optimal how small of fault signature based on intelligent optimization algorithm
Ripple basic function；
Step 3, the signal gathered using optimal multi-wavelet bases function pair carry out redundancy multi-wavelet transformation, after obtaining multiple decomposition
Sub-band；
Step 4, relative energy ratio at the fault characteristic frequency of each sub-band is calculated, by the higher sub-band of relative energy
As the sensitive sub-band where combined failure feature；
Step 5, Hilbert envelope demodulation processing is carried out to the sensitive sub-band that decomposition obtains one by one, extract epicyclic gearbox
Combined failure correlated characteristic, and carry out identifying and diagnosing.
2. a kind of epicyclic gearbox combined failure feature extracting method according to claim 1, it is characterized in that, step 2.1
In, free parameter is determined by solving the Lifting Coefficients non trivial solution owed under fixed condition.
3. a kind of epicyclic gearbox combined failure feature extracting method according to claim 1, it is characterized in that, step 2.2
Detailed process it is as follows：
First, morther wavelet of the selection with suitable vanishing moment, determines partition function, and describe partition function by quality factor
Scaling behavior, and then Legendre is carried out to quality factor and converts to obtain the multifractal spectra of signal；
Secondly, comentropy is calculated and be introduced into multifractal spectra, construct multi-fractal entropy, and be normalized to obtain normalizing
Change multi-fractal entropy；
Finally, based on intelligent optimization algorithm, the normalization of detail signal corresponding to latter two wavelet basis function is decomposed with m ultiwavelet
The minimum optimization object function of multi-fractal entropy sum optimizes, and obtains adaptive multi-wavelet bases function.
4. a kind of epicyclic gearbox combined failure feature extracting method according to claim 1 or 3, it is characterized in that, step
In 2.2, intelligent optimization algorithm is genetic algorithm.
5. a kind of epicyclic gearbox combined failure feature extracting method according to claim 1 or 3, it is characterized in that, step
In 2.2, multi-fractal entropy H is normalizednFor：
<mrow>
<msub>
<mi>H</mi>
<mi>n</mi>
</msub>
<mo>=</mo>
<mo>-</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>k</mi>
</munderover>
<mfrac>
<mrow>
<mi>f</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>&alpha;</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>k</mi>
</munderover>
<mi>f</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>&alpha;</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
</mrow>
</mfrac>
<msub>
<mi>log</mi>
<mi>k</mi>
</msub>
<mrow>
<mo>(</mo>
<mfrac>
<mrow>
<mi>f</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>&alpha;</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>k</mi>
</munderover>
<mi>f</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>&alpha;</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
</mrow>
</mfrac>
<mo>)</mo>
</mrow>
</mrow>
In formula, f (αi) be signal multifractal spectra.
6. a kind of epicyclic gearbox combined failure feature extracting method according to claim 1, it is characterized in that, in step 3,
The number of plies that redundancy multi-wavelet transformation is carried out to multi-wavelet bases function is 3 layers.
7. a kind of epicyclic gearbox combined failure feature extracting method according to claim 1, it is characterized in that, in step 4,
Each the relative energy at the fault characteristic frequency of sub-band is than r：
Wherein fc∈(fc- Δ, fc+Δ)
In formula, A be squared envelope spectrum amplitude, fcFrequency is characterized, Δ is selected frequency separation, and f=0~f ' is son frequency
Band scope.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710868167.3A CN107608936B (en) | 2017-09-22 | 2017-09-22 | Method for extracting compound fault characteristics of planetary gear box |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710868167.3A CN107608936B (en) | 2017-09-22 | 2017-09-22 | Method for extracting compound fault characteristics of planetary gear box |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107608936A true CN107608936A (en) | 2018-01-19 |
| CN107608936B CN107608936B (en) | 2020-09-15 |
Family
ID=61057567
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710868167.3A Active CN107608936B (en) | 2017-09-22 | 2017-09-22 | Method for extracting compound fault characteristics of planetary gear box |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107608936B (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109443768A (en) * | 2018-11-14 | 2019-03-08 | 中国直升机设计研究所 | A kind of Helicopter Main Reducer planetary gear vibration signal separation method |
| CN110660057A (en) * | 2019-11-01 | 2020-01-07 | 重庆大学 | Binocular automatic gear pitting detection device based on deep learning |
| CN110844111A (en) * | 2019-10-11 | 2020-02-28 | 中国直升机设计研究所 | Multi-characteristic index bevel gear health state assessment method |
| CN110850230A (en) * | 2019-11-07 | 2020-02-28 | 天津大学 | Method for extracting interpolar short circuit fault characteristic quantity of direct current distribution network based on multi-wavelet theory |
| CN112633121A (en) * | 2020-12-18 | 2021-04-09 | 国网浙江省电力有限公司武义县供电公司 | Radiation source identification method based on Hilbert transform and multi-fractal dimension characteristics |
| CN112665851A (en) * | 2020-10-28 | 2021-04-16 | 西安交通大学 | Key-free phase change rotating speed gearbox fault diagnosis method |
| ES2754278R1 (en) * | 2018-09-28 | 2021-07-01 | Univ Navarra Publica | FAULT DIAGNOSIS IN PLANETARY GEARS |
| CN113688904A (en) * | 2021-08-24 | 2021-11-23 | 大连海事大学 | Intelligent ship system equipment dynamic characteristic parameter extraction method |
| CN113740055A (en) * | 2021-07-14 | 2021-12-03 | 西安交通大学 | Method and device for separating and diagnosing composite fault components of gear box |
| CN117515131A (en) * | 2024-01-04 | 2024-02-06 | 之江实验室 | Method, device, storage medium and equipment for monitoring abrasion of planetary reducer |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101804465A (en) * | 2010-03-31 | 2010-08-18 | 桂林电子科技大学 | Multi-scale wavelet antivibration design method of high-speed main shaft of machine tool |
| CN102879196A (en) * | 2012-09-25 | 2013-01-16 | 西安交通大学 | Compound fault diagnosing method for planetary gearbox by using matrix wavelet transformation |
| CN103954353A (en) * | 2014-05-12 | 2014-07-30 | 苏州大学 | Signal transient element sparse representation detection method and device and failure detection method |
| CN105445022A (en) * | 2015-11-17 | 2016-03-30 | 中国矿业大学 | Planetary gear fault diagnosis method based on dual-tree complex wavelet transform-entropy feature fusion |
| US20160124031A1 (en) * | 2014-11-04 | 2016-05-05 | Walid G. Morsi Ibrahim | Smart multi-purpose monitoring system using wavelet design and machine learning for smart grid applications |
| CN106404399A (en) * | 2016-11-03 | 2017-02-15 | 北京印刷学院 | Bearing fault diagnosis method based on adaptive redundancy lifting wavelet packet decomposition tree |
-
2017
- 2017-09-22 CN CN201710868167.3A patent/CN107608936B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101804465A (en) * | 2010-03-31 | 2010-08-18 | 桂林电子科技大学 | Multi-scale wavelet antivibration design method of high-speed main shaft of machine tool |
| CN102879196A (en) * | 2012-09-25 | 2013-01-16 | 西安交通大学 | Compound fault diagnosing method for planetary gearbox by using matrix wavelet transformation |
| CN103954353A (en) * | 2014-05-12 | 2014-07-30 | 苏州大学 | Signal transient element sparse representation detection method and device and failure detection method |
| US20160124031A1 (en) * | 2014-11-04 | 2016-05-05 | Walid G. Morsi Ibrahim | Smart multi-purpose monitoring system using wavelet design and machine learning for smart grid applications |
| CN105445022A (en) * | 2015-11-17 | 2016-03-30 | 中国矿业大学 | Planetary gear fault diagnosis method based on dual-tree complex wavelet transform-entropy feature fusion |
| CN106404399A (en) * | 2016-11-03 | 2017-02-15 | 北京印刷学院 | Bearing fault diagnosis method based on adaptive redundancy lifting wavelet packet decomposition tree |
Non-Patent Citations (1)
| Title |
|---|
| 申婷婷: "《基于小波分析的齿轮箱故障诊断技术研究》", 《CNKI》 * |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| ES2754278R1 (en) * | 2018-09-28 | 2021-07-01 | Univ Navarra Publica | FAULT DIAGNOSIS IN PLANETARY GEARS |
| CN109443768A (en) * | 2018-11-14 | 2019-03-08 | 中国直升机设计研究所 | A kind of Helicopter Main Reducer planetary gear vibration signal separation method |
| CN110844111A (en) * | 2019-10-11 | 2020-02-28 | 中国直升机设计研究所 | Multi-characteristic index bevel gear health state assessment method |
| CN110844111B (en) * | 2019-10-11 | 2022-11-04 | 中国直升机设计研究所 | Multi-characteristic index bevel gear health state assessment method |
| CN110660057A (en) * | 2019-11-01 | 2020-01-07 | 重庆大学 | Binocular automatic gear pitting detection device based on deep learning |
| CN110850230A (en) * | 2019-11-07 | 2020-02-28 | 天津大学 | Method for extracting interpolar short circuit fault characteristic quantity of direct current distribution network based on multi-wavelet theory |
| CN110850230B (en) * | 2019-11-07 | 2022-04-29 | 天津大学 | Method for extracting interpolar short circuit fault characteristic quantity of direct current distribution network based on multi-wavelet theory |
| CN112665851A (en) * | 2020-10-28 | 2021-04-16 | 西安交通大学 | Key-free phase change rotating speed gearbox fault diagnosis method |
| CN112633121A (en) * | 2020-12-18 | 2021-04-09 | 国网浙江省电力有限公司武义县供电公司 | Radiation source identification method based on Hilbert transform and multi-fractal dimension characteristics |
| CN113740055A (en) * | 2021-07-14 | 2021-12-03 | 西安交通大学 | Method and device for separating and diagnosing composite fault components of gear box |
| CN113740055B (en) * | 2021-07-14 | 2022-08-09 | 西安交通大学 | Method and device for separating and diagnosing composite fault components of gear box |
| CN113688904A (en) * | 2021-08-24 | 2021-11-23 | 大连海事大学 | Intelligent ship system equipment dynamic characteristic parameter extraction method |
| CN113688904B (en) * | 2021-08-24 | 2023-08-22 | 大连海事大学 | Method for extracting dynamic characteristic parameters of intelligent ship system equipment |
| CN117515131A (en) * | 2024-01-04 | 2024-02-06 | 之江实验室 | Method, device, storage medium and equipment for monitoring abrasion of planetary reducer |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107608936B (en) | 2020-09-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107608936A (en) | A kind of epicyclic gearbox combined failure feature extracting method | |
| CN102879196B (en) | Compound fault diagnosing method for planetary gearbox by using matrix wavelet transformation | |
| CN102937522B (en) | Composite fault diagnosis method and system of gear case | |
| CN104483127B (en) | A kind of planetary gear feature information of weak faults extracting method | |
| CN102721545B (en) | Rolling bearing failure diagnostic method based on multi-characteristic parameter | |
| Liu et al. | Acoustic emission analysis for wind turbine blade bearing fault detection under time-varying low-speed and heavy blade load conditions | |
| CN103439110A (en) | Early-stage weak fault diagnosis method of rolling bearing | |
| CN110132578B (en) | Compound fault feature extraction method for gear system and fault test device | |
| CN102607845A (en) | Bearing fault characteristic extracting method for redundantly lifting wavelet transform based on self-adaptive fitting | |
| CN110991419B (en) | Gearbox local fault diagnosis method based on sparse low-rank collaborative optimization framework | |
| CN108801630A (en) | The gear failure diagnosing method of single channel blind source separating | |
| CN109765052B (en) | Planetary gearbox early fault diagnosis method based on GOA-ASR | |
| CN104374575A (en) | Wind turbine main bearing fault diagnosis method based on blind source separation | |
| CN102721537B (en) | Mechanical impact type fault diagnosis method based on variable space-scale frame | |
| CN110046476A (en) | The ternary two of rolling bearing fault is into the sparse diagnostic method of Fractal Wavelet | |
| Zhang et al. | A novel compound fault diagnosis method using intrinsic component filtering | |
| CN104330258A (en) | Method for identifying grey relational degree of rolling bearing fault based on characteristic parameters | |
| CN110333071B (en) | Mechanical vibration signal processing method using narrowband cepstrum transformation | |
| CN106404386A (en) | Method for acquiring, extracting and diagnosing early stage fault characteristic signals of gear case | |
| CN107886078A (en) | A kind of Threshold Denoising method based on layered self-adapting threshold function table | |
| CN102879195A (en) | Multi-wavelet self-adaptive block threshold noise-reducing time domain diagnosis method for damage of gearbox | |
| CN115434872A (en) | Wind turbine generator gearbox composite fault diagnosis method based on AVMD and improved RSSD | |
| CN109946076B (en) | Planetary wheel bearing fault identification method of weighted multi-scale dictionary learning framework | |
| CN115034263A (en) | Variable-scale evolution self-adaptive noise elimination method, noise elimination system and fault diagnosis system | |
| Li et al. | Incipient detection of bearing fault using impulse feature enhanced weighted sparse representation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |